GeoCFNet: Geometry-Aware Confidence Field Network for Robot-Assisted Endoscopic Submucosal Dissection

2026-06-11 · Source: Computer Vision and Pattern Recognition · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems, Computer Vision · Depth: Expert, quick

Summary

GeoCFNet is a geometry-aware confidence field network designed to provide stable and precise visual guidance for robot-assisted endoscopic submucosal dissection (ESD). ESD, a promising approach for en-bloc resection of large lesions, faces challenges in reliable confidence field estimation due to dynamic endoscopic scenes, including smoke, specular highlights, tissue deformation, and weak texture. GeoCFNet addresses this by formulating dissection guidance as a geometry-aware confidence field estimation problem. Built on a pretrained DINOv3 backbone, the network integrates a Token-Differentiated Fusion module to aggregate class-token context with dense patch representations, a SegFormer decoder for confidence regression, and Geometry-Aware Spatial Regularization (GASR) to preserve spatial coherence and local geometric transitions. Experimental results demonstrate accurate and geometrically stable confidence field estimation, achieving RMSE 0.0480, PSNR 27.1995, SSIM 0.3397, and CC 0.2466.

Key takeaway

For AI Scientists developing visual guidance systems for robot-assisted endoscopic submucosal dissection, GeoCFNet presents a robust framework for accurate confidence field estimation. Its integration of geometry-aware spatial regularization and a Token-Differentiated Fusion module effectively addresses challenges like dynamic scenes and weak tissue texture. You should evaluate incorporating similar geometry-aware principles and advanced feature aggregation techniques to enhance the stability and precision of your surgical guidance models.

Key insights

GeoCFNet enhances robot-assisted ESD guidance via geometry-aware confidence field estimation.

Principles

• Dense confidence fields effectively represent dissection regions and spatial transitions.
• Geometry-aware formulation is crucial for reliable confidence field estimation.

Method

GeoCFNet integrates a Token-Differentiated Fusion module, a SegFormer decoder for confidence regression, and Geometry-Aware Spatial Regularization (GASR) on a DINOv3 backbone.

In practice

• Improve visual guidance in robot-assisted ESD.
• Enhance safety margins during endoscopic dissection.

Topics

• Robot-assisted Surgery
• Endoscopic Submucosal Dissection
• Confidence Field Estimation
• GeoCFNet
• Computer Vision
• DINOv3

Best for: Computer Vision Engineer, AI Scientist, Machine Learning Engineer, Research Scientist

Related on AIssential

• Point-Wise Geometry-Aware Transformer for Partial-to-Full Point Cloud Registration in Computer-Assisted Surgery — GAPR-Net fuses local and global geometry via a transformer and invariant features for robust partial-to-full point cloud registration.
• Towards Learning a Generalizable 3D Scene Representation from 2D Observations — A Generalizable Neural Radiance Field predicts 3D workspace occupancy from robot observations in a global frame.
• Computational Imaging Priors for Wireless Capsule Endoscopy: Monte Carlo-Guided Hemoglobin Mapping for Rare-Anomaly Detection — A Monte Carlo-inspired prior improves wireless capsule endoscopy classification by better distinguishing hemoglobin from confounding factors.
• GeoSAM-3D: Geodesic Prompt Propagation for Open-Vocabulary 3D Scene Segmentation from Monocular Video — GeoSAM-3D propagates 3D segmentation prompts using geodesic heat-kernel distance on Gaussian splatting scenes from monocular video, improving mask continuity and reducing leakage.
• Canonical Regularisation of Wide Feature-Learning Neural Networks — Traditional ridge regularization biases feature-learning networks due to curved parameter-space geometry, necessitating a geodesic-based approach.
• A Data-Centric Framework for Intraoperative Fluorescence Lifetime Imaging for Glioma Surgical Guidance — A data-centric AI framework improves FLIm accuracy for glioma surgery by refining labels and enhancing model robustness.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computer Vision and Pattern Recognition.